Primary Anorexia Nervosa or Weight Phobia in the Male:Report on I3 Cases

A. H. CRISP, D. A. TOMS

British Medical Journal, 1972, 1, 334-338

SummaryThe cases of 13 men with anorexia nervosa are reported.While the disorder as seen in the clinic is much lesscommon in males than females this may not be a truereflection of the differential sex or overall prevalencerates. The disorder is found to have the same basiccharacteristics in the male as in the female: namely, aphobic avoidance of normal weight associated withelective carbohydrate starvation and emaciation. As inthe female the mechanism appears to develop out ofnormal adolescent dieting behaviour and to arise as a

means of avoiding overwhelming psychosocial matura-tional demands of adolescence. Premorbid and familyfeatures include a state ofovernutrition and a high degreeof family psychopathology reflected in high rates ofparental psychiatric morbidity and marital difficulty.

Introduction

Primary anorexia nervosa in the female is thought to be rare.This conclusion is probably greatly affected by a central aspectof the disorder-the subject's reluctance or intense fear ofchangeand her insistence that she is not ill. Many therefore avoidbecoming patients, even sometimes to the point of death, andothers who are precipitated into medical care successfullyconceal the true basis of their symptomatology and behaviour.Nevertheless, patients from the background with the greatestprevalence of the disorder-namely, adolescent girls fromsocial classes I and II-will sometimes convincingly acknow-ledge that at any one time between one in 50 and one in 100of their fellow schoolgirls display a similar severe and seeminglyindisputable form of the disorder. The disorder may be becom-ing more common, but this is hard to establish.

Series of 50 or more women have now been carefully docu-mented and described by several clinical research workers.All agree that the disorder is much more rare in the male(Crisp, 1967; Dally, 1969; Theander, 1970), reports of thesex ratio ranging from 1 in 10 to 1 in 20. Bliss and Branch (1960)suggested that the male prevalence rate may be greater thanthis but that it is masked by a greater reluctance on the partof clinicians to diagnose the condition in young men. In par-ticular, secondary amenorrhoea cannot draw attention to theneed for further inquiry into the psychological, social, andnutritional history of the individual. Dally (1969) reviewedsome of the scanty literature on male anorexia nervosa anddescribed six more cases. Bruch (1971) added another five cases,which she carefully distinguished from four other male patientswith anorexia and cachexia associated with other psychiatricillnesses and without the central characteristics of primaryanorexia nervosa. These she identified as refusal to eat pre-dominantly carbohydrate foods in the pathological pursuit ofthinness together with major body-image and somatic-perceptualdisturbances.

Several authors have drawn attention to the presence of

possibly significant homosexual conflict preceding the onset ofthe disorder in up to 50% of their male patients (Crisp, 1967;Dally, 1969). Clinical features of premorbid obesity and laterdrug and alcohol dependence and "compulsive" overeatinghave been emphasized in several reported case histories. Dally(1969) concluded that the disorder carried a worse prognosisin the male. In all about 50 cases have been described in theliterature over the past two centuries in more or less detail.

This report concerns the cases of 11 male patients studied indetail over the past 10 years together with two other patientsincluded after attention had been drawn to their case historiesbecause of their family association with women with primaryanorexia nervosa receiving treatment under our care. All thesubjects were considered to have severe and definite primaryanorexia nervosa-namely, a state of emaciation associatedwith inability to ingest or retain ingested food and characteris-tically accompanied by low basal metabolism, daytime andnocturnal restlessness, and preoccupation with food. Associatedwith the inability to eat was a resistance to eating fatteningfoods, often arising from an established fear of fatness, butalways linked with a progressively single-minded pursuit ofthinness associated with a developing fear of and determinationto avoid normal adolescent and adult weight. This fear and theconsequent feeding disorder had usually been concealed ordenied at first.

Seven of the patients received a period of inpatient careduring which their weight was restored to the level of theirmatched population mean weight, while at the same time theyand their families were given psychotherapy. The four otherpatients were seen and treated as outpatients. Details of theclinical presentation and course of some of these cases and theother two subjects are given elsewhere (Crisp and Roberts,1962; Crisp, 1967; Toms and Crisp, 1972). The data presentedhere have to do with some of the factors previously found to beimportant in female anorexia nervosa (Crisp, 1970). Theycomprise (a) some social, family, and clinical characteristicsof the total group, and (b) results of the investigation duringtreatment of one patient whose distorted growth we were ableto examine in some detail.

MethodInformation about psychological and social aspects of patientsand especially about their personal histories is notoriouslyinaccurate. This report reflects part of a larger investigation intosuch aspects in female patients with primary anorexia nervosa,and throughout we have tried to improve the quality of thedata by using as many sources as possible. This information wasderived from the patients and their families, usually over aprolonged period, and concerns the background of the subjectsand their families, their clinical state, and the course of thedisorder. In some instances the subjects have been comparedwith subgroups or the total group in 160 cases in female patientsseen and documented (Crisp, 1970) by A. H. C. over the same

period.

Family Background

SOCIAL CLASS

The parental social class in Case 12 was not recorded. Sixof the other 12 subjects came from a social class I or II back-

St. George's Hospital Medical School, St. George's Hospital, LondonS.W.17

A. H. CRISP, M.D., M.R.C.P., Professor of PsychiatryD. A. TOMS, M.R.C.P., D.P.M., Senior Registrar in Psychiatry

334 BRITISH MEDICAL jouRNAL 5 FEBRUARY 1972

on 30 Septem

ber 2020 by guest. Protected by copyright.

http://ww

w.bm

j.com/

Br M

ed J: first published as 10.1136/bmj.1.5796.334 on 5 F

ebruary 1972. Dow

nloaded from

BRITISH MEDICAL JOURNAL 5 FEBRUARY 1972

ground and the other six from a social class III, IV, or V back-ground (father's occupation, Registrar General's classification).This can be compared with the even greater predominance ofsocial class I and II backgrounds (40 among the first 55 subjects)in the parallel series of female patients, but this differencedoes not reach the 5% level of significance.

BIRTH ORDER

The reported birth order in 12 of the subjects is shown in TableI. Seven were first-born and five were second-born. When thisfinding is compared with the reported birth order of the first55 subjects in the female series (22 first-born, 22 second-born,7 born third or later, and 4 unknown) the tendency for the malesto be more often first-born is seen not to reach statistical sig-nificance.

TABLE i-Birth Order in 13 Male Subjects with Primary Anorexia Nervosaor Weight Phobia (Reported Live Births)

Case No. Birth Order

1 Brother 2 Sister2~~~~~~~~~fSister

2 1 { Sister3 Sister 24 Sister 25 1 Sister6 1 Brother Sister7 18 Sister 29 Sister 210 1 Brother* Brother11 1 Brother Sister*12 ?13 1 Sister* Brother

* Sibling with anorexia nervosa.

PARENTAL AGE

The mean paternal age was 31 years at the time of the probands'birth and the mean maternal age 29 years. The standard devia-tions were small.

FAMILY NUTRITIONAL HISTORY

The weight and shape of most of the parents was not abnormal.Several parents were intensely preoccupied with their physicalfitness and with maintaining normal weight, and in associationwith this several mothers and fathers severely restricted theircarbohydrate intake either consistently or sporadically. Ofthree fathers working with food, however, two were chefs andchronically obese.The occurrence of primary anorexia nervosa among other

members of the family was remarkable. From Table I it is seenthat three siblings had had the condition. In Cases 11 and 13the disorder was in the sister and was carefully documented,in one instance by ourselves. In Case 10 the disorder had almostcertainly occurred in the second-born brother. In these threeinstances the disorder developed in sequence between thesiblings concerned. In Case 10 first the eldest brother (underour care) developed the condition. As he recovered his brotherdeveloped the disorder. In Case 11 it is likely that the elderbrother (under our care when he reached the age of 26 years)had developed the disorder first. During what proved to be aremission in his condition his younger sister developed thedisorder, which lasted about three years. As she recovered heredeveloped the disorder in a severe form, which persisted.In Case 13 the occurrence of the disorder in the elder brotherwas brought to our attention by the presentation with it of hissister. She had been ill for about three years after his recovery.

Finally, Case 12 is included in this series mainly because wethink the facts of this case provide evidence for the importanceof environmental background factors in the disorder. He was

a very thin 46-year-old man who presented as the adoptivefather of a 14-year-old patient with primary anorexia nervosa.He was himself financially successful as a sweet-food manufac-turer. It was evident that he had been the subject of primaryanorexia nervosa for many years. His marriage was stable butinfertile and possibly unconsummated, and it had resulted inadoption of a baby who was by now the adolescent patientpresenting with the disorder. A second girl with primaryanorexia nervosa subsequently presented to A.H.C., and oneof the features in her history was a period in early adolescencewhen she had stayed with this family as an evacuee during thesecond world war. This period had ended about five yearsbefore the onset of her anorexia nervosa; she had becomeprogressively obese at that time.

PSYCHIATRIC HISTORY IN THE FAMILY

There was a broad but not necessarily remarkable scatter ofpsychiatric illness in the wider family circles of the probands.Nevertheless, apart from frequent marital interactional diffi-culties there was also a substantial amount of major psychiatricillness and personality disorder among the parents. One fatherwas a severe alcoholic, another had crippling chronic agora-phobia associated with years of psychiatric care in hospital.One mother had killed herself after years of chronic anxietyand depression, a second had a chronic paranoid psychosis, athird had a severe chronic state of anxiety depression with avariety of phobic avoidance patterns, and a fourth had a historyof agoraphobia and depressive illness requiring admission tohospital and electric convulsion therapy. Thus there wasevidence that a quarter of the 24 fathers and mothers had orhad had formal severe and sometimes chronic psychiatricillness, not infrequently involving social phobic avoidancepatterns.The family conflicts, which we believe to be important

contributory factors, took the usual variety of forms (Crisp,1970) and were in our view linked with the patients' psycho-pathology, including their abnormal attitudes to food as partof the resolution of their adolescent difficulties.

Patients' History and IllnessCLINICAL PREVALENCE

The 11 male patients and two other subjects were seen over a10-year period during which a series of 160 female patientswith primary anorexia nervosa were also referred for diagnosisand treatment. This gives a differential sex prevalence for these10 years in the clinic of one male to 15 females.

BIRTH WEIGHT AND DURATION OF BREAST-FEEDING

These data were available on 10 subjects. The mean reportedbirth weight was 7 lb 10 oz, S.D. 13 oz (3,450 g, S.D. 370 g).The reported duration of breast-feeding varied from up tothree months for eight of the subjects to five and 12 monthsrespectively for the remaining two.

BEHAVIOUR AND NUTRITION IN CHILDHOOD

This information was available only at the anecdotal descriptivelevel. As children the probands were reported as having had avariety of neurotic symptoms which neither characterizedthem as a group nor clearly distinguished them from otherchildren. Several were described as having been "plumpish"but this was not invariable. Most were said to have had goodappetites, usually without a history of food fads or other nurtur-ent reflections of mother-child conflict. The parents also tended

335

on 30 Septem

ber 2020 by guest. Protected by copyright.

http://ww

w.bm

j.com/

Br M

ed J: first published as 10.1136/bmj.1.5796.334 on 5 F

ebruary 1972. Dow

nloaded from

336

to look back nostalgically on the probands' compliant natures aschildren.

WEIGHT JUST BEFORE ONSET OF DISORDER

These weights expressed as a percentage of matched populationmean weights are shown in Fig. 1. It is clear that the probandswere an overweight population at this point.

vg L.El

UC

-4

, 3D ^,

-0

E 2

Z

Presenting weight

mean = 77.50/oS.D. ±11 5

( difference = 36- 5%)

0/0m.p.m.W.

z

25 C

0

10 7 3

I I 2 '

13 6

1201101090 8070 60 50 40

FIG. 1-Premorbid and presenting weights in 12 males.Results given as percentage of matched population meanweight (% m.p.m.w.) for corresponding ages.

ASPECTS OF CLINICAL STATUS

The adolescent age of onset of the disorder and the probands'later age of presentation to doctors are shown in Fig. 2. Thepresenting weight of each subject, again expressed as a per-

centage of matched population mean weight, is given in Fig. 1.

4

&-.'v 2

0

E 2

z 34

Age at onsetmean =16b2 yearsS.D.=+ 32

11 12 13 14 15 16 17 18 19D 21 22 23 24 25 26 27 28 29 30 31 32(year)- A at( rsentati

Age at presentationmean =198 years

S.D.= ±61(mean difference =3b6years)

FIG. 2-Reported age of onset and presentation in 13 males withprimary anorexia nervosa or weight phobia.

TABLE II-Feeding Habits in the 13 Male Patients

Feeding ActivityCase AgeNo. Abstain Binge- Vomit- Purgimng mgin ig Alcohol or Drugs

(A) (B) (V) (P)1 14 A2 17 B Vt3 16 B Vt P*4 20 A5 21 A V Alcohol6 22 A B*7 15 A P* Alcohol8 15 B V*t Smoked pot9 34 A BB P Alcohol10 19 A B11 26 A B Occasional

alcohol abuse12 46 A13 16 A

*Severe.tElectrolyte disturbance: hypokalsemia.

BRITISH MEDICALJOURNAL 5 FEBRUARY 1972

The feeding patterns are shown in Table II. Patterns ofchronic vomiting and purging were predictably associated withsevere electrolyte disturbance, most strikingly reflected in severehypokalemia. Symptomatic epilepsy, based on such metabolicdisturbances, was a feature of one and possibly two cases,and was reminiscent of a similar prevalence in the femalepopulation (Crisp et al., 1968). Other aspects of behaviourincluded typical preoccupation with body weight, diets, andcooking; general hyperactivity sometimes incorporated intoritualistic behaviour, nocturnal restlessness, propensities fordrug dependence-for example, cigarette smoking and theother drug ingestions noted in TableII-sometimes embarkedon in a conscious effort to stave off bulimia and consequentweight gain by stifling appetite and ingesting what was assumedto be a substance of less calorific value, and at other times as amore complex social action; and the absence of sexual activityat either the behavioural or fantasy level.

Apart from the subject in Case 12 another patient (Case 9)had been married. His marriage, established after the onsetof the disorder, had been unconsummated and had ended indivorce after five years. The general mood state varied betweensubjects and was also highly variable in the individual. Noneshowed sustained deep depression or anxiety, but they tendedto describe sequential feelings such as panics, elation, anguish,and irritability set against the background of restlessness and,when severely ill, fatigue and exhaustion. At times they wouldtry to assume and convey a manner of blandness, reflecting theiracceptance of their present state and their wish not to invokethe concern and attempts of others to change them. They were,however, also able, under certain circumstances, to describetheir paralysingly mixed feelings which required them toremain as they were, realizing the hopelessness of their presentposition but never able to take the fearful step of eating theirway to "recovery."

TREATMENT AND OUTCOME

This report was not written to dwell on individual psycho-pathology except in so far as its origins are reflected in therelatively "hard" data described above. We believe, however,that we were able to identify overwhelming individual matura-tional problems in all our patients. These were reflected in thepatients' histories as a variety of sexual conflicts and identitycrises, often arising within the context of the parental andmarital conflicts briefly referred to above.

Subjects tended to be reluctant and sometimes resistant aspatients. The 11 who presented as patients had done so onlyafter being "cornered," usually by their general practitionersafter attention had been drawn to some superficial but seriousaspect of their disorder, such as severe bulimia, vomitingand diarrhoea, symptomatic epilepsy, emaciation, bizarrefood fads, or impotence. In these circumstances their complaintstook such forms as their desire to eat but inability to do sobecause of consequent abdominal discomfort or vomiting.We felt justified in effecting the compulsory detention of onepatient who was grossly emaciated and weak at presentationbut refused to accept treatment.

Six other patients agreed to hospital admission and specifictreatment aimed at restoring their weight to their individualmatched population mean levels, together with normal dietaryintake, as a basis for attempting to help them with the emotionaldifficulties provoked by such change. One of these cases wasreported in detail elsewhere (Crisp and Roberts, 1962). Theother six also regained weight precisely in the way describedabove. Only two patients, however, managed to maintain thisand began to explore and cope with their re-emerging matura-tional problems. Two of the others became bulimic and morealcoholic, as they had feared, and one started regular vomitingfor the first time. There was a tendency among two or threeof the subjects for antisocial behaviour to re-emerge, which

I lTI

I

11

F1F-,I 1- I I , - F,-,-I -,-7

on 30 Septem

ber 2020 by guest. Protected by copyright.

http://ww

w.bm

j.com/

Br M

ed J: first published as 10.1136/bmj.1.5796.334 on 5 F

ebruary 1972. Dow

nloaded from

BRITISH MEDICAL JOURNAL 5 FEBRUARY 1972

had not been evident throughout the duration of the disorderbut -had sometimes threatened or appeared briefly before inearly adolescence.The case of one patient (Case 7) is reported here in greater

detail because the disorder, which started very early in hispuberty, clearly stunted his growth, including his stature. Itis shown in Fig. 3 that his height remained stationary duringthe years 12-15 inclusive. After the initial treatment successfully

80 -actual position

70 50 chronoloqical aqe 97/18 position expected with/ uninterrupted growth

SO cLLC7C

bO~~~~~~~~~~~~~~~~~~/

05

40

~~ ~ ~ ~ ~ ~ L-, '

35 40 45 50 55 60 65 70 75Heiqht (inches)

FIG. 3-Case 7. Weight for height with percentiles (Tanner,1958).

restored his weight to normal he began to grow in height,picking up his growth curve again during the next 18 months.By the end of this period, however, he had lapsed into a mixedstate of delinquent behaviour and social phobic avoidancebehaviour. He became unmanageable in the general hospitalpsychiatric unit and was admitted to a psychiatric hospital withthe predominant symptom of agoraphobia together with heavyalcohol consumption-the former being the same manifest dis-order as his father had experienced since adolescence.

Discussion

In our view primary anorexia nervosa in the male is probablya rare condition though it mnay be much more common thanis evident in the clinic. In our experience male patients tendto be even more resistant to treatment than females, and even

if some features of their disorder are brought to medical atten-tion the fundamental psychopathological abnormalities mayremain concealed while the patient allows himself to be investi-gated for gastrointestinal and endocrine dysfunction or thedepressive or obsessional features of his mental state.The disorder in the male has the same central morbidity,

reflecting itself in a phobic avoidance of normal adult weightand attendant impulses and expectations and the consequentsocial complications and maturational demands. As such it isalso a primary disorder of weight and not primarily a feedingdisorder. As in the female the adolescent neurotic conflictswhich precipitate the disorder appear to be in no way specificto it. In other subjects such conflicts might- either be dealt witheffectively or promote other disorders of adolescence. We regardthe high prevalence of psychiatric illness in the families ofthese patients, which has been commented on before in relationto female patients with anorexia nervosa (Kay and Leigh, 1954),as another sign of this general underlying neurotic morbidity.The predisposition to develop primary anorexia nervosa in

such circumstances is in our view governed by a variety of factors,among which the major one is dieting behaviour, itself strikinglyless common in adolescent males than females.

Such behaviour, often started in an effort to reshape appear-ance in pursuit of greater attractiveness and self-confidenceand a securer identity, may be a response to problems which

337

subsequently lead direct to overdetermination of the dietingbehaviour and the development of primary anorexia nervosa.At other times such preliminary dieting appears to have beenat first a more independent step. In all instances, however, itappears to provide the initial mechanism of entry into thecondition of primary anorexia nervosa, determining the latter'sdevelopment as the "chosen" defensive posture.Such behaviour in the cases described occurred against a

background of overweight, sometimes though not alwayssevere, and about which the subjects had been sensitive andsometimes teased. Such premorbid overweight often reflectsa long-standing state which in females has been found to beassociated with faster-than-average growth rate both in thegeneral population and in the population with anorexia nervosa(Crisp, 1970). If this is also true for males it nevertheless remainstrue that, overall, boys reach puberty some year or so later thangirls. Such relative protection from precipitation into adolescenceat a very early age may be another factor contributing to thedifferential sex incidence in the disorder.

It became evident during inquiry and treatment that most ofthe present male patients had during their early years beenoverprotected, especially by their mothers. Such concern hadoften also involved attempts at overnurturing since the time ofbirth or even conception. These maternal needs, tending to occurin relatively old primiparae, were based on personal or maternalinsecurities and anxieties, the ultimate expression of which wassometimes the frank maternal psychiatric disorders referred toabove. Not too much can be made of any specific aspects ofpaternal morbidity, which was more varied in this small series.It was noted, however, that three of the fathers worked withfood. Three were contributing acutely to the family instabilitythrough their current virtual defections from the marriage.Already as children, then, the patients appeared ill-equipped toenter into and cope with the maturational demands of adoles-cence and possibly also likely to invoke nurturent means ofdealing with such problems. In the three cases in which morethan one sibling had had the disocder it appeared to us that theywere acting, so to speak, in series, needing in the process tomaintain some family interaction and immobilize some threat-ened change.The ultimate family basis for the nurturent factors described

above probably derives from a complex interaction of geneticand experiential influences. The case of the adoptive family iscited to show the probable important contribution of the latter.The major effect that the disorder can have on physical

growth, especially when its onset is early in adolescence, isshown in Case 7. This finding is in accord with the reportedtendency for patients in the parallel female population in whomthe disorder has begun during or before the 17th year to beshorter in stature than a population matched for sex, age, andsocial class. Such retardation and stunting of growth is probablyrelated to malnutrition consequent on predominant carbo-hydrate starvation but other depletions may be relevant,especially in those subjects given to chronic and often covertvomiting and purging. It is also noteworthy that the abovepatient complained of severe insomnia; this has been shownelsewhere to be a feature of primary anorexia nervosa, as hasa reduction in R.E.M. sleep (Crisp et al., 1971). It has beenshown (Sassin et al., 1969) that there is a complex but probablyoverall direct relation between such forms of sleep and produc-tion of growth hormone.

Finally, with this small series it was not thought worth whileto compare the course of the disorder in the males with thatin the females. Nevertheless, it was our impression that thisspecific male patient population carried a worse prognosis thanthe parallel female population.

We would like to acknowledge our gratitude to the large numberof other people also concerned in the care of these patients,especially the nursing staff at Atkinson Morley's Hospital andMrs. B. Harding, psychiatric social worker.

on 30 Septem

ber 2020 by guest. Protected by copyright.

http://ww

w.bm

j.com/

Br M

ed J: first published as 10.1136/bmj.1.5796.334 on 5 F

ebruary 1972. Dow

nloaded from

338 BRITISH MEDICAL JOURNAL 5 FEBRUARY 1972

ReferencesBliss, E. L., and Branch, C. H. H. (1960). Anorexia Nervosa. New York,

Hoeber.Bruch, H. (1971). Psychosomatic Medicine, 33, 31.Crisp, A. H. (1967). Hospital Medicine, 1, 713.Crisp, A. H. (1970). World Review of Nutrition and Dietetics, 12, 452.Crisp, A. H., Fenton, G. W., and Scotton, L. (1968). British Journal of

Psychiatry, 114, 1149.Crisp, A. H., and Roberts, F. J. (1962). Postgraduate Medical Journal, 38,

350.

Crisp, A. H., Stonehill, E., and Fenton, G. W. (1971). Postgraduate MedicalJournal, 47, 207.

Daily, P. (1969). Anorexia Nervosa. London, Heinemann.Kay, D. W. K., and Leigh, D. (1954). Journal of Mental Science, 100, 411.Sassin, J. F., et al. (1969). Science, 165, 513.Tanner, J. M. (1958). In Modern Trends in Paediatrics ed. A. Holzel and J.

Tizard, p. 331. London, Butterworths.Theander, S. (1970). Acta Psychiatrica Scandinavica, Suppl. No. 214.Toms, D. A., and Crisp, A. H. (1972). Journal of Psychosomatic Research, 16.

In press.

Effects of Renal Function on Plasma Digoxin Levels inElderly Ambulant Patients in Domiciliary Practice

E. MARY BAYLIS, M. S. HALL, GILLIAN LEWIS, VINCENT MARKS

British Medical Journal, 1972, 1, 338-341

Summary

An investigation into the relations between the dailydose of digoxin, drug regimen, serum digoxin concentra-tion, and creatinine and digoxin clearance was carriedout in a group ofelderly ambulant patients in domiciliarypractice. Moderate to severe impairment of renal func-tion was found both in patients taking digoxin and inelderly control subjects. Plasma digoxin levels were notrelated to blood urea concentrations or creatinineclearance. Digoxin clearance was less than creatinineclearance. Now that plasma digoxin levels can be meas-ured relatively easily their estimation should becomepart of clinical practice.

Introduction

Despite nearly 200 years of clinical experience since the publica-tion of William Withering's An Account of the Foxglove andSome of its Medical Uses, difficulties are still encountered inprescribing cardiac glycosides. Toxic symptoms are oftenproduced in one patient with doses therapeutically effective foranother. With the advent of radioimmunoassay techniques(Smith et al., 1969) for measuring serum digoxin concentrationsit has been appreciated that this may be largely due to overlapbetween toxic and therapeutic blood levels (Chamberlain et al.,1970; Smith and Haber, 1970), although individual variation insensitivity to the glycosides is important, and other factors-including potassium (Sampson et al., 1943), calcium (Moe andFarah, 1965), and magnesium (Seller et al., 1970) concentra-tions, sodium balance (Harrison and Wakim, 1969), thyroidstatus (Doherty, 1968), the presence of severe heart disease, orchronic pulmonary disease (Beller et al., 1971)-influence theliability to develop toxicity.From 67 to 77% of the digoxin is present in the blood in the

protein-free (Evered et al., 1970) pharmacologically active(Lullman and Van Swieten, 1969) state, and its concentration

Epsom Hospital Laboratories, West Park Hospital, Epsom, SurreyE. MARY BAYLIS, B.M., B.CH., Senior Registrar in Chemical PathologyGILLIAN LEWIS, B.SC., Biochemist

Forest Row, SussexM. S. HALL, M.B., B.S., D.OBST.R.C.O.G., General Practitioner

University of Surrey, Guildford, SurreyVINCENT MARKS, D.M., F.R.C.P., Professor of Clinical Biochemistry

is to a great extent a function of its rate of renal clearance.The tendency for elderly patients to require smaller doses ofdigoxin to achieve adequate digitalization or therapeutic bloodlevels (Dall, 1965; Doherty, 1968; Ewy, et al., 1969; Chamber-lain et al., 1970) and the poor correlation between serum levelsand the daily dose in the presence of impaired renal function(Chamberlain et al., 1970) have been noted, suggesting thatrenal function should be assessed before prescribing digoxin.In the studies reported (Smith et al., 1969; Chamberlain et al.,1970; Smith and Haber, 1970; Beller et al., 1971) renal functionwas judged by measuring the blood urea concentration, anotoriously late and insensitive index of renal disease whichhas moreover been shown on occasion to be well within "normallimits" when the renal digoxin clearance was seriously reduced(Doherty et al., 1967).More valuable information is provided by measuring the

creatinine clearance, and this has been used as a method ofadjusting digitalis therapy before blood measurements becameavailable (Jelliffe and Blankenhom, 1967). Ideally, the serumdigoxin concentration should be maintained at a more or lessconstant, effective level throughout the day, and more accurateknowledge of renal function will influence the dose and mode ofadministration. Paediatric doses, 0-065 mg, may be useful insome cases. This report describes the results of an investigationinto the relations between the daily dose of digoxin, drugregimen, serum digoxin concentration, and creatinine clearancein a group of elderly ambulant patients in general practice.

Patients and Methods

Thirty-one randomly selected subjects who had clinically beenwell controlled by a constant dose of digoxin for at least onemonth were investigated. Blood samples were taken when thesepatients presented at the surgery, either spontaneously (usuallyto obtain a repeat digoxin prescription) or by request. Clinicaldetails recorded in each case include: name, age, date of birth,sex; nature of the heart condition; other significant diseasespresent; drugs; digoxin therapy (with details of dosage, durationof therapy, frequency of administration, and exact time relationsbetween venepuncture and taking the previous dose); andhistory of symptoms which could be due to digoxin therapy.

Second and sometimes third blood samples were collectedfrom most patients at a prearranged time of day to gain furtherinformation about the plasma digoxin level at varying intervalsafter taking the drug. In 25 patients creatinine clearances weremeasured. Each patient was given precise verbal and writteninstructions how to collect a 24-hour sample of urine and wasvisited during that period for blood sampling. (Only one blood

on 30 Septem

ber 2020 by guest. Protected by copyright.

http://ww

w.bm

j.com/

Br M

ed J: first published as 10.1136/bmj.1.5796.334 on 5 F

ebruary 1972. Dow

nloaded from